{[ promptMessage ]}

Bookmark it

{[ promptMessage ]}

224s.09.lec14 - CS224S/LING281 SpeechRecognition,Synthesis...

Info icon This preview shows pages 1–12. Sign up to view the full content.

View Full Document Right Arrow Icon
CS 224S/LING 281  Speech Recognition, Synthesis,  and Dialogue Dan Jurafsky Lecture 14:  Dialogue: MDPs  and Speaker Detection
Image of page 1

Info icon This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon
Outline for today MDP Dialogue Architectures Speaker Recognition
Image of page 2
Now that we have a success metric Could we use it to help drive learning? In recent work we use this metric to help  us learn an optimal  policy  or  strategy  for  how the conversational agent should  behave
Image of page 3

Info icon This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon
New Idea: Modeling a dialogue system  as a probabilistic agent A conversational agent can be  characterized by: The current knowledge of the system A set of  states   S  the agent can be in a set of  actions   A  the agent can take goal   G , which implies A success metric that tells us how well the agent  achieved its goal A way of using this metric to create a strategy or  policy   π  for what action to take in any particular  state.
Image of page 4
What do we mean by actions A and  policies  π ? Kinds of decisions a conversational agent  needs to make: When should I ground/confirm/reject/ask for  clarification on what the user just said? When should I ask a directive prompt, when an open  prompt? When should I use user, system, or mixed initiative?
Image of page 5

Info icon This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon
A threshold is a human-designed  policy! Could we learn what the right action is Rejection Explicit confirmation Implicit confirmation No confirmation By learning a policy which,  given various information about the current  state, dynamically chooses the action which  maximizes dialogue success
Image of page 6
Another strategy decision Open versus directive prompts When to do mixed initiative How we do this optimization? Markov Decision Processes
Image of page 7

Info icon This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon
Review: Open vs. Directive Prompts Open prompt System gives user very few constraints User can respond how they please: “How may I help you?” “How may I direct your  call?” Directive prompt Explicit instructs user how to respond “Say yes if you accept the call; otherwise, say  no”
Image of page 8
Review: Restrictive vs. Non- restrictive gramamrs Restrictive grammar Language model which strongly constrains  the ASR system, based on dialogue state Non-restrictive grammar Open language model which is not restricted  to a particular dialogue state
Image of page 9

Info icon This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon
Kinds of  Initiative How do I decide which of these initiatives  to use at each point in the dialogue? Grammar Open Prompt Directive Prompt Restrictive Doesn’t make sense System Initiative Non-restrictive User Initiative Mixed Initiative
Image of page 10
Modeling a dialogue system as a  probabilistic agent A conversational agent can be  characterized by: The current knowledge of the system A set of 
Image of page 11

Info icon This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon
Image of page 12
This is the end of the preview. Sign up to access the rest of the document.

{[ snackBarMessage ]}

What students are saying

  • Left Quote Icon

    As a current student on this bumpy collegiate pathway, I stumbled upon Course Hero, where I can find study resources for nearly all my courses, get online help from tutors 24/7, and even share my old projects, papers, and lecture notes with other students.

    Student Picture

    Kiran Temple University Fox School of Business ‘17, Course Hero Intern

  • Left Quote Icon

    I cannot even describe how much Course Hero helped me this summer. It’s truly become something I can always rely on and help me. In the end, I was not only able to survive summer classes, but I was able to thrive thanks to Course Hero.

    Student Picture

    Dana University of Pennsylvania ‘17, Course Hero Intern

  • Left Quote Icon

    The ability to access any university’s resources through Course Hero proved invaluable in my case. I was behind on Tulane coursework and actually used UCLA’s materials to help me move forward and get everything together on time.

    Student Picture

    Jill Tulane University ‘16, Course Hero Intern